Vaccine candidates targeting clade B virus currently in human clinical trials are slated for Phase 3 testing at the Anonymous Testing Center at the Hospital Escola So Francisco of the Federal University of Rio de Janeiro, Brazil, a key international site of the HIV Vaccine Trials Network (HVTN) and the site for the current proposal. Vaccine efficacy will be measured by decreased rates of HIV infection among vaccine compared to placebo recipients. Since trial sample size is a function of the underlying incidence of new infections among the volunteer population, characterizing a risk profile of those at highest risk for trial enrollment facilitates trial efficiency. Andre P[unreadable]riss[unreadable], MD, MSc a Brazilian long-term AITRP trainee proposes to investigate the hypothesis that the application of ego network theory-based approaches that characterize the ego-defined profile of the subject's personal network will provide additional information for characterizing volunteer HIV risk and thus augment selection of high risk subjects for participation in HIV phase 3 vaccine trials. Aim 1 will apply ego-network-based techniques to profile persons at high risk for HIV infection in order to create an Ego Risk Factor Index based on ego's personal risk profile and ego's report of network contact risk factors. Among previously untested volunteers this cross-sectional study will investigate the degree to which Ego Risk Factor Index predicts HIV status. Aim 2 will compare patterns of risk behavior 5- years and l-year prior to study enrollment to determine the stability of Ego Risk Factor Index over time as a means of validating the assumption that ego-defined HIV risk profile precedes HIV infection. Employing a cross-sectional design 1250 volunteers will be recruited with an underlying prevalence of 8% in order enroll 100 HIV positive and 1150 negative volunteers. Applying a previously validated ego-network questionnaire adapted to the Brazilian context, each individual will be interviewed about his/her own risk factors and asked to report all sexual contacts and their risk factors during the past 5 years. The Ego Risk Factor Index will be calculated based on logistic regression models fitted with variables related to ego and several indices created for each risk factor information collected from all sexual partners presented in the neighborhood o fan ego's personal network. This project will facilitate future HIV vaccine trials by creating a new tool for risk assessment and build sophisticated capacity for ego-network research in Brazil. The study will be performed in Brazil as an extension of NIH Grant # U01-AI48001, 61112000-513112005. [unreadable] [unreadable]